install cuda

Peter Caspers

30 Mar 2013 30 Mar '13

3:43 p.m.

Hi, I am trying to install the cuda package on a Windows 7 enviroment. However I run into an error and can not figure out, what it is. Can someone help ? Thanks a lot Peter C:\Users\Peter\_dev\haskell>cabal install cuda Resolving dependencies... [1 of 1] Compiling Main ( C:\Users\Peter\AppData\Local\Temp\cuda-0.5.0.1-11856\cuda-0.5.0.1\Setup.hs, C:\Users\Peter\AppData\Local\Temp\cuda-0.5.0.1 -11856\cuda-0.5.0.1\dist\setup\Main.o ) Linking C:\Users\Peter\AppData\Local\Temp\cuda-0.5.0.1-11856\cuda-0.5.0.1\dist\setup\setup.exe ... Configuring cuda-0.5.0.1... cygwin warning: MS-DOS style path detected: C:\Users\Peter\AppData\Roaming\cabal/share/config.site Preferred POSIX equivalent is: /cygdrive/c/Users/Peter/AppData/Roaming/cabal/share/config.site CYGWIN environment variable option "nodosfilewarning" turns off this warning. Consult the user's guide for more details about POSIX paths: http://cygwin.com/cygwin-ug-net/using.html#using-pathnames checking for gcc... C:\Program Files (x86)\Haskell Platform\2012.4.0.0\mingw\bin\gcc.exe checking whether the C compiler works... no configure: error: in `/cygdrive/c/Users/Peter/AppData/Local/Temp/cuda-0.5.0.1-11856/cuda-0.5.0.1': configure: error: C compiler cannot create executables See `config.log' for more details cabal: Error: some packages failed to install: cuda-0.5.0.1 failed during the configure step. The exception was: ExitFailure 77

Show replies by date

Henk-Jan van Tuyl

30 Mar 30 Mar

7:56 p.m.

On Sat, 30 Mar 2013 20:43:58 +0100, Peter Caspers wrote:

...

Hi,

I am trying to install the cuda package on a Windows 7 enviroment. However I run into an error and can not figure out, what it is. : : configure: error: C compiler cannot create executables See `config.log' for more details :

The message says it all: See `config.log' for more details The config.log file is probably in the directory where cabal-install unpacks it (%appdata%\cabal\cuda-<revision>); you can also use the commands: cabal unpack cuda cd cuda cabal install notepad config.log I just had the same error message for another package; from the config.log file it became clear, that the linker could not find the necessary library. Make sure, that the libraries and header files can be found by the compiler and linker, by setting the proper environment variables, see [0] Regards, Henk-Jan van Tuyl [0] http://www.haskell.org/haskellwiki/Windows#Tools_for_compilation -- http://Van.Tuyl.eu/ http://members.chello.nl/hjgtuyl/tourdemonad.html Haskell programming --

Trevor L. McDonell

10:01 p.m.

Hi, CUDA package maintainer here. I don't have access to a Win7 box with compatible GPU, nor a lot of experience writing packages for windows. If you do get it working please send me a pull request on github. I'd be great to have this working for windows as well. Cheers, -Trevor On 31/03/2013, at 10:56 AM, Henk-Jan van Tuyl wrote:

...

On Sat, 30 Mar 2013 20:43:58 +0100, Peter Caspers wrote:

...
Hi,

I am trying to install the cuda package on a Windows 7 enviroment. However I run into an error and can not figure out, what it is. : : configure: error: C compiler cannot create executables See `config.log' for more details :

The message says it all: See `config.log' for more details The config.log file is probably in the directory where cabal-install unpacks it (%appdata%\cabal\cuda-<revision>); you can also use the commands: cabal unpack cuda cd cuda cabal install notepad config.log

I just had the same error message for another package; from the config.log file it became clear, that the linker could not find the necessary library. Make sure, that the libraries and header files can be found by the compiler and linker, by setting the proper environment variables, see [0]

Regards, Henk-Jan van Tuyl

[0] http://www.haskell.org/haskellwiki/Windows#Tools_for_compilation

-- http://Van.Tuyl.eu/ http://members.chello.nl/hjgtuyl/tourdemonad.html Haskell programming --

_______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe

Peter Caspers

31 Mar 31 Mar

11:03 a.m.

Hi, thank you. I could resolve some of the problems by removing spaces from the Cuda and Haskell platform installation paths. Now I am left wiht the following error: configure:3596: checking for library containing cuDriverGetVersion configure:3627: c:\HaskellPlatform\2012.4.0.0\mingw\bin\gcc.exe -o conftest.exe -Wl,--hash-size=31 -Wl,--reduce-memory-overheads -I/c/CUDA/NVIDIA_GPU_Computing_Toolkit/CUDA/v4.1/include -L/c/CUDA/NVIDIA_GPU_Computing_Toolkit/CUDA/v4.1/lib conftest.c >&5 C:\Users\Peter\AppData\Local\Temp\ccOsnsjD.o:conftest.c:(.text+0xc): undefined reference to `cuDriverGetVersion' collect2: ld returned 1 exit status In fact the library path -L/c/CUDA/NVIDIA_GPU_Computing_Toolkit/CUDA/v4.1/lib is not correct (there are two subfolders x64 and Win32 containing the lib files) and I do not see where this path is actually taken from. I defined an enviroment variable LD_LIBRARY_PATH with the correct paths (separated by a colon), but it seems to be ignored. Also copying the lib files to the path I see in the log does not help. Do you have any hint for me concerning this ? Peter Am 31.03.2013 00:56, schrieb Henk-Jan van Tuyl:

...

On Sat, 30 Mar 2013 20:43:58 +0100, Peter Caspers wrote:

...
Hi,

I am trying to install the cuda package on a Windows 7 enviroment. However I run into an error and can not figure out, what it is. : : configure: error: C compiler cannot create executables See `config.log' for more details :

The message says it all: See `config.log' for more details The config.log file is probably in the directory where cabal-install unpacks it (%appdata%\cabal\cuda-<revision>); you can also use the commands: cabal unpack cuda cd cuda cabal install notepad config.log

I just had the same error message for another package; from the config.log file it became clear, that the linker could not find the necessary library. Make sure, that the libraries and header files can be found by the compiler and linker, by setting the proper environment variables, see [0]

Regards, Henk-Jan van Tuyl

[0] http://www.haskell.org/haskellwiki/Windows#Tools_for_compilation

Henk-Jan van Tuyl

11:59 a.m.

On Sun, 31 Mar 2013 17:03:22 +0200, Peter Caspers wrote:

...

In fact the library path -L/c/CUDA/NVIDIA_GPU_Computing_Toolkit/CUDA/v4.1/lib is not correct (there are two subfolders x64 and Win32 containing the lib files) and I do not see where this path is actually taken from. I defined an enviroment variable LD_LIBRARY_PATH with the correct paths (separated by a colon), but it seems to be ignored.

Also copying the lib files to the path I see in the log does not help.

The environment variable should probably be LIBRARY_PATH; I use a semicolon as separator. See also LD_LIBRARY_PATH vs LIBRARY_PATH[0]. Regards, Henk-Jan van Tuyl [0] http://stackoverflow.com/questions/4250624/ld-library-path-vs-library-path -- http://Van.Tuyl.eu/ http://members.chello.nl/hjgtuyl/tourdemonad.html Haskell programming --

Peter Caspers

12:55 p.m.

...

The environment variable should probably be LIBRARY_PATH; I use a semicolon as separator. See also LD_LIBRARY_PATH vs LIBRARY_PATH[0].

yes, it's LIBRARY_PATH. The x64 version of cuda.lib is not recognized at all (same error message as if the file was not existent). The Win32 version "works", but results in configure:3627: c:\HaskellPlatform\2012.4.0.0\mingw\bin\gcc.exe -o conftest.exe -Wl,--hash-size=31 -Wl,--reduce-memory-overheads -I/c/CUDA/NVIDIA_GPU_Computing_Toolkit/CUDA/v4.1/include -L/c/CUDA/NVIDIA_GPU_Computing_Toolkit/CUDA/v4.1/lib conftest.c -lcuda >&5 C:\Users\Peter\AppData\Local\Temp\ccOwCQ6n.o:conftest.c:(.text+0xc): undefined reference to `cuDriverGetVersion' collect2: ld returned 1 exit status I ran nm on cuda.lib and got the entry nvcuda.dll: 00000000 I .idata$4 00000000 I .idata$5 00000000 I .idata$6 00000000 T .text U _IMPORT_DESCRIPTOR_nvcuda 00000000 I _imp__cuDriverGetVersion@4 00000000 T cuDriverGetVersion@4 this looks ok so far. Running nm on the x64 version of the lib file results in rubbish output (consistent with the observation above). I understand that LD_LIBRARY_PATH is used to look up to dll when running the program (is that correct?). However we are not at this point yet, are we, since the error occurs on the gcc invocation ?

...

Try my fork:

https://github.com/mainland/cuda

In particular, read WINDOWS.md.

I also read Geoffreys WINDOWS.md and understood that configuring dll names are only necessary when using ghci, not for compiled programs (nothing to do for this case ?) and in particular not for installing the package ? Actually the dll is not named nvcuda.dll as indicated in the nm output, but rather cudart32_41_28.dll I suppose and this file is located in the bin subfolder. I should set LD_LIBRARY_PATH to the bin folder, yes ? Should I configure this dll name for package installation already (i.e. in addition to what is mentioned in WINDOWS.md) ? If yes, how ? Thank you Peter

Geoffrey Mainland

2:17 p.m.

On 03/31/2013 05:55 PM, Peter Caspers wrote:

...

...
The environment variable should probably be LIBRARY_PATH; I use a

semicolon as separator.

...
See also LD_LIBRARY_PATH vs LIBRARY_PATH[0].

yes, it's LIBRARY_PATH. The x64 version of cuda.lib is not recognized at all (same error message as if the file was not existent). The Win32 version "works", but results in

configure:3627: c:\HaskellPlatform\2012.4.0.0\mingw\bin\gcc.exe -o conftest.exe -Wl,--hash-size=31 -Wl,--reduce-memory-overheads -I/c/CUDA/NVIDIA_GPU_Computing_Toolkit/CUDA/v4.1/include -L/c/CUDA/NVIDIA_GPU_Computing_Toolkit/CUDA/v4.1/lib conftest.c -lcuda >&5 C:\Users\Peter\AppData\Local\Temp\ccOwCQ6n.o:conftest.c:(.text+0xc): undefined reference to `cuDriverGetVersion' collect2: ld returned 1 exit status

I ran nm on cuda.lib and got the entry

nvcuda.dll: 00000000 I .idata$4 00000000 I .idata$5 00000000 I .idata$6 00000000 T .text U _IMPORT_DESCRIPTOR_nvcuda 00000000 I _imp__cuDriverGetVersion@4 00000000 T cuDriverGetVersion@4

this looks ok so far. Running nm on the x64 version of the lib file results in rubbish output (consistent with the observation above).

You're using a version of GHC that targets 32-bit x86, so the 64-bit library is not going to do you any good. cuDriverGetVersion@4 is the stdcall-mangled version of cuDriverGetVersion. The CUDA headers (at least in the 5.0 toolkit) do not properly declare all CUDA functions as stdcall under gcc, even though they are. This is why you get an undefined reference---gcc see cuDriverGetVersion' declared ccall in the header, so it looks for the wrong symbol. I would guess that the 4.1 toolkit has this same problem the same. In general, there is no support for using the CUDA SDK with the mingw tools. I hacked around this enough to get it to work for the driver API, but you have to use my fork.

...

I understand that LD_LIBRARY_PATH is used to look up to dll when running the program (is that correct?). However we are not at this point yet, are we, since the error occurs on the gcc invocation ?

...
Try my fork:

https://github.com/mainland/cuda

In particular, read WINDOWS.md.

I also read Geoffreys WINDOWS.md and understood that configuring dll names are only necessary when using ghci, not for compiled programs (nothing to do for this case ?) and in particular not for installing the package ?

There's more to it than reading my WINDOWS.md. If you want to build the cuda package at all, you also need to use my fork.

...

Actually the dll is not named nvcuda.dll as indicated in the nm output, but rather cudart32_41_28.dll I suppose and this file is located in the bin subfolder. I should set LD_LIBRARY_PATH to the bin folder, yes ? Should I configure this dll name for package installation already (i.e. in addition to what is mentioned in WINDOWS.md) ? If yes, how ?

There are two dlls you need. nvcuda.dll corresponds to libcuda on Linux, and its .lib file on Windows is cuda.lib. cudart32_41_28.dll corresponds to libcudart on Linux, and its .lib file on Windows is cudart.lib. You will need them both. I was able to install the cuda package under 32-bit GHC 7.4.2 using the 5.0 SDK and use it from within ghci. This required using my fork of the cuda repo and following the instructions in my WINDOWS.md. Make sure nvcc is in your path (the CUDA installer should have made this so) and try 'cabal configure'. LD_LIBRARY_PATH is used on most UNIX flavors. I don't believe it does anything in Windows. You will instead need to modify your Path environment variable from the System Settings->Advanced System Settings control panel. Geoff

Peter Caspers

2:32 p.m.

yes I more or less saw this in the meantime, too. Actually modifying the source code on which the error is reported from configure: failed program was: | /* confdefs.h */ | #define PACKAGE_NAME "Haskell CUDA bindings" | #define PACKAGE_TARNAME "cuda" ... | #ifdef __cplusplus | extern "C" | #endif | char cuDriverGetVersion (); | int | main () | { | return cuDriverGetVersion (); | ; | return 0; | } to (for example) #include #include #include #include int main( int argc, char** argv) { int driverVersion = 0; cuDriverGetVersion(&driverVersion); printf("version = %d\n",driverVersion); return 0; } let me compile, link and run without errors. Alright, got it now, I will try your github. Do you think it works with CUDA 4.1 ? On my laptop this is the latest version that runs due to the card driver. Thanks in any case Peter

Peter Caspers

3:27 p.m.

Hmm, I get Configuring cuda-0.5.0.0... setup.exe: configure script not found. can you help ? Peter

...

I was able to install the cuda package under 32-bit GHC 7.4.2 using the 5.0 SDK and use it from within ghci. This required using my fork of the cuda repo and following the instructions in my WINDOWS.md. Make sure nvcc is in your path (the CUDA installer should have made this so) and try 'cabal configure'.

Geoffrey Mainland

3:48 p.m.

You need to generate the configure script using autoconf: https://www.gnu.org/software/autoconf/manual/autoconf.html#autoconf-Invocati... On 03/31/2013 08:27 PM, Peter Caspers wrote:

...

Hmm, I get

Configuring cuda-0.5.0.0... setup.exe: configure script not found.

can you help ?

Peter

...
I was able to install the cuda package under 32-bit GHC 7.4.2 using the 5.0 SDK and use it from within ghci. This required using my fork of the cuda repo and following the instructions in my WINDOWS.md. Make sure nvcc is in your path (the CUDA installer should have made this so) and try 'cabal configure'.

Peter Caspers

1 Apr 1 Apr

5:33 a.m.

yes, the installation seems to work fine now. However, doing the following test in ghci Prelude> :m +Foreign.CUDA Prelude Foreign.CUDA> props 0 Loading package bytestring-0.9.2.1 ... linking ... done. Loading package cuda-0.5.0.0 ... linking ... done. results in a crash. The CUDA version I am using is 4.1.28. You think there is something I could try to analyze this further ? Thanks a lot for your help Peter Am 31.03.2013 21:48, schrieb Geoffrey Mainland:

...

You need to generate the configure script using autoconf:

https://www.gnu.org/software/autoconf/manual/autoconf.html#autoconf-Invocati...

On 03/31/2013 08:27 PM, Peter Caspers wrote:

...
Hmm, I get

Configuring cuda-0.5.0.0... setup.exe: configure script not found.

can you help ?

Peter

...
I was able to install the cuda package under 32-bit GHC 7.4.2 using the 5.0 SDK and use it from within ghci. This required using my fork of the cuda repo and following the instructions in my WINDOWS.md. Make sure nvcc is in your path (the CUDA installer should have made this so) and try 'cabal configure'.

Geoffrey Mainland

6:25 a.m.

...

yes, the installation seems to work fine now. However, doing the following test in ghci

Prelude> :m +Foreign.CUDA Prelude Foreign.CUDA> props 0 Loading package bytestring-0.9.2.1 ... linking ... done. Loading package cuda-0.5.0.0 ... linking ... done.

results in a crash. The CUDA version I am using is 4.1.28. You think

That is not a very elucidating crash message, so I don't see how to proceed. After ghci print "Loading package cuda-0.5.0.0 ... linking ... done." it just exits? No error dialog, nothing? Did you try building any of the examples in the cuda package that don't require ghci? Is your graphics card incompatible with CUDA 5.0, or do you just not want to update your driver? Geoff On 04/01/2013 10:33 AM, Peter Caspers wrote: there is something I could try to analyze this further ?

...

Thanks a lot for your help Peter

Am 31.03.2013 21:48, schrieb Geoffrey Mainland:

...
You need to generate the configure script using autoconf:

https://www.gnu.org/software/autoconf/manual/autoconf.html#autoconf-Invocati...

...

...
On 03/31/2013 08:27 PM, Peter Caspers wrote:

...
Hmm, I get

Configuring cuda-0.5.0.0... setup.exe: configure script not found.

can you help ?

Peter

...
I was able to install the cuda package under 32-bit GHC 7.4.2 using the 5.0 SDK and use it from within ghci. This required using my fork of the cuda repo and following the instructions in my WINDOWS.md. Make sure nvcc is in your path (the CUDA installer should have made this so) and try 'cabal configure'.

Peter Caspers

11:27 a.m.

indeed, not very helpful ... When I installed Cuda the latest driver (296.0) that was running on my laptop (a W520 ThinkPad) was not sufficient for version 5.0. However as I noticed today in February Lenovo released a driver update (311.0) and with that 5.0 is in fact running. :-) With that the Haskell bindings work well. Thanks again very much, Geoff Peter Am 01.04.2013 12:25, schrieb Geoffrey Mainland:

...

That is not a very elucidating crash message, so I don't see how to proceed. After ghci print "Loading package cuda-0.5.0.0 ... linking ... done." it just exits? No error dialog, nothing? Did you try building any of the examples in the cuda package that don't require ghci?

Is your graphics card incompatible with CUDA 5.0, or do you just not want to update your driver?

Geoff

...
yes, the installation seems to work fine now. However, doing the following test in ghci

Prelude> :m +Foreign.CUDA Prelude Foreign.CUDA> props 0 Loading package bytestring-0.9.2.1 ... linking ... done. Loading package cuda-0.5.0.0 ... linking ... done.

results in a crash. The CUDA version I am using is 4.1.28. You think

On 04/01/2013 10:33 AM, Peter Caspers wrote: there is something I could try to analyze this further ?

...
Thanks a lot for your help Peter

Am 31.03.2013 21:48, schrieb Geoffrey Mainland:

...
You need to generate the configure script using autoconf:

https://www.gnu.org/software/autoconf/manual/autoconf.html#autoconf-Invocati...

...
On 03/31/2013 08:27 PM, Peter Caspers wrote:

...
Hmm, I get

Configuring cuda-0.5.0.0... setup.exe: configure script not found.

can you help ?

Peter

...
I was able to install the cuda package under 32-bit GHC 7.4.2 using the 5.0 SDK and use it from within ghci. This required using my fork of the cuda repo and following the instructions in my WINDOWS.md. Make sure nvcc is in your path (the CUDA installer should have made this so) and try 'cabal configure'.

Geoffrey Mainland

12:01 p.m.

Fantastic, glad you got it working! Maybe it's time for me to send Trevor a pull request... Geoff On 04/01/2013 04:27 PM, Peter Caspers wrote:

...

indeed, not very helpful ...

When I installed Cuda the latest driver (296.0) that was running on my laptop (a W520 ThinkPad) was not sufficient for version 5.0. However as I noticed today in February Lenovo released a driver update (311.0) and with that 5.0 is in fact running. :-)

With that the Haskell bindings work well.

Thanks again very much, Geoff Peter

Am 01.04.2013 12:25, schrieb Geoffrey Mainland:

...
That is not a very elucidating crash message, so I don't see how to proceed. After ghci print "Loading package cuda-0.5.0.0 ... linking ... done." it just exits? No error dialog, nothing? Did you try building any of the examples in the cuda package that don't require ghci?

Is your graphics card incompatible with CUDA 5.0, or do you just not want to update your driver?

Geoff

...
yes, the installation seems to work fine now. However, doing the following test in ghci

Prelude> :m +Foreign.CUDA Prelude Foreign.CUDA> props 0 Loading package bytestring-0.9.2.1 ... linking ... done. Loading package cuda-0.5.0.0 ... linking ... done.

results in a crash. The CUDA version I am using is 4.1.28. You think

On 04/01/2013 10:33 AM, Peter Caspers wrote: there is something I could try to analyze this further ?

...
Thanks a lot for your help Peter

Am 31.03.2013 21:48, schrieb Geoffrey Mainland:

...
You need to generate the configure script using autoconf:

https://www.gnu.org/software/autoconf/manual/autoconf.html#autoconf-Invocati...

...
...
...
On 03/31/2013 08:27 PM, Peter Caspers wrote:

...
Hmm, I get

Configuring cuda-0.5.0.0... setup.exe: configure script not found.

can you help ?

Peter

...
I was able to install the cuda package under 32-bit GHC 7.4.2 using the 5.0 SDK and use it from within ghci. This required using my fork of the cuda repo and following the instructions in my WINDOWS.md. Make sure nvcc is in your path (the CUDA installer should have made this so) and try 'cabal configure'.

Manuel M T Chakravarty

3 Apr 3 Apr

10:05 p.m.

Geoffrey Mainland :

...

Fantastic, glad you got it working! Maybe it's time for me to send Trevor a pull request...

That sounds like an excellent idea! Manuel

...

On 04/01/2013 04:27 PM, Peter Caspers wrote:

...
indeed, not very helpful ...

When I installed Cuda the latest driver (296.0) that was running on my laptop (a W520 ThinkPad) was not sufficient for version 5.0. However as I noticed today in February Lenovo released a driver update (311.0) and with that 5.0 is in fact running. :-)

With that the Haskell bindings work well.

Thanks again very much, Geoff Peter

Am 01.04.2013 12:25, schrieb Geoffrey Mainland:

...
That is not a very elucidating crash message, so I don't see how to proceed. After ghci print "Loading package cuda-0.5.0.0 ... linking ... done." it just exits? No error dialog, nothing? Did you try building any of the examples in the cuda package that don't require ghci?

Is your graphics card incompatible with CUDA 5.0, or do you just not want to update your driver?

Geoff

...
yes, the installation seems to work fine now. However, doing the following test in ghci

Prelude> :m +Foreign.CUDA Prelude Foreign.CUDA> props 0 Loading package bytestring-0.9.2.1 ... linking ... done. Loading package cuda-0.5.0.0 ... linking ... done.

results in a crash. The CUDA version I am using is 4.1.28. You think

On 04/01/2013 10:33 AM, Peter Caspers wrote: there is something I could try to analyze this further ?

...
Thanks a lot for your help Peter

Am 31.03.2013 21:48, schrieb Geoffrey Mainland:

...
You need to generate the configure script using autoconf:

https://www.gnu.org/software/autoconf/manual/autoconf.html#autoconf-Invocati...

...
...
...
On 03/31/2013 08:27 PM, Peter Caspers wrote:

...
Hmm, I get

Configuring cuda-0.5.0.0... setup.exe: configure script not found.

can you help ?

Peter

> I was able to install the cuda package under 32-bit GHC 7.4.2 using the > 5.0 SDK and use it from within ghci. This required using my fork of the > cuda repo and following the instructions in my WINDOWS.md. Make sure > nvcc is in your path (the CUDA installer should have made this so) and > try 'cabal configure'.

_______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe

Trevor L. McDonell

10:37 p.m.

Hi Geoff, Yes, please do! -T On 02/04/2013, at 3:01 AM, Geoffrey Mainland wrote:

...

Fantastic, glad you got it working! Maybe it's time for me to send Trevor a pull request...

Geoff

On 04/01/2013 04:27 PM, Peter Caspers wrote:

...
indeed, not very helpful ...

When I installed Cuda the latest driver (296.0) that was running on my laptop (a W520 ThinkPad) was not sufficient for version 5.0. However as I noticed today in February Lenovo released a driver update (311.0) and with that 5.0 is in fact running. :-)

With that the Haskell bindings work well.

Thanks again very much, Geoff Peter

Am 01.04.2013 12:25, schrieb Geoffrey Mainland:

...
That is not a very elucidating crash message, so I don't see how to proceed. After ghci print "Loading package cuda-0.5.0.0 ... linking ... done." it just exits? No error dialog, nothing? Did you try building any of the examples in the cuda package that don't require ghci?

Is your graphics card incompatible with CUDA 5.0, or do you just not want to update your driver?

Geoff

...
yes, the installation seems to work fine now. However, doing the following test in ghci

Prelude> :m +Foreign.CUDA Prelude Foreign.CUDA> props 0 Loading package bytestring-0.9.2.1 ... linking ... done. Loading package cuda-0.5.0.0 ... linking ... done.

results in a crash. The CUDA version I am using is 4.1.28. You think

On 04/01/2013 10:33 AM, Peter Caspers wrote: there is something I could try to analyze this further ?

...
Thanks a lot for your help Peter

Am 31.03.2013 21:48, schrieb Geoffrey Mainland:

...
You need to generate the configure script using autoconf:

https://www.gnu.org/software/autoconf/manual/autoconf.html#autoconf-Invocati...

...
...
...
On 03/31/2013 08:27 PM, Peter Caspers wrote:

...
Hmm, I get

Configuring cuda-0.5.0.0... setup.exe: configure script not found.

can you help ?

Peter

> I was able to install the cuda package under 32-bit GHC 7.4.2 using the > 5.0 SDK and use it from within ghci. This required using my fork of the > cuda repo and following the instructions in my WINDOWS.md. Make sure > nvcc is in your path (the CUDA installer should have made this so) and > try 'cabal configure'.

_______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe

Stephen Tetley

31 Mar 31 Mar

2:54 a.m.

It looks like you are using Cygwin for a Unix-alike environment. For building Haskell bindings to C libraries you are better off with MinGW + MSYS. On 30 March 2013 19:43, Peter Caspers wrote:

...

I am trying to install the cuda package on a Windows 7 enviroment. However I run into an error and can not figure out, what it is.

Can someone help ?

Geoffrey Mainland

12:23 p.m.

Try my fork: https://github.com/mainland/cuda In particular, read WINDOWS.md. Geoff On 03/31/2013 07:54 AM, Stephen Tetley wrote:

...

It looks like you are using Cygwin for a Unix-alike environment. For building Haskell bindings to C libraries you are better off with MinGW + MSYS.

On 30 March 2013 19:43, Peter Caspers wrote:

...
I am trying to install the cuda package on a Windows 7 enviroment. However I run into an error and can not figure out, what it is.

Can someone help ?

4430

Age (days ago)

4435

Last active (days ago)

List overview

Download

17 comments

6 participants

participants (6)

Geoffrey Mainland
Henk-Jan van Tuyl
Manuel M T Chakravarty
Peter Caspers
Stephen Tetley
Trevor L. McDonell

install cuda

Peter Caspers

Trevor L. McDonell

Peter Caspers

Peter Caspers

Peter Caspers

Peter Caspers

Peter Caspers

Peter Caspers

Trevor L. McDonell

tags

participants (6)